Protocol Online logo
Top : Forum Archives: : DNA Methylation, Histone and Chromatin Study

Please help me analyze BSP cloned sequencing results. - (Mar/19/2008 )

Dear all, I used Chemicon CpGenome Universal Modification kit to treat DNA. The DNA-treated was then subject to PCR and product was sent to company for T-A cloning and sequencing. I have aligned the cloned sequence and C-T conversion sequence from original target sequence. But there is no significant similarity. I am also very curious that no G is present in the cloned sequence. However, both forward and reverse primer can match to the cloned sequence, and the length of PCR product is as expected. Please find sequences below attached. Please help me analyze this clonee sequence results and reasons of such results. I will appreciate your every suggestions and comments.

Original target sequence

1738-cccctctgctagtgtgacacacttcgcgcaactccgcagttggcgggcgcggaccacccctgcggctctgccggctggct
gtcaccctcgggggctctggctgccgacccacggggcgggctcccgagcggttccaagcctcggagctgggcgggggcgg
gagggagcctgggagaa-1914

Sequence after C-T conversion from original target sequence

1738-tttttttgttagtgtgatatatttcgcgtaatttcgtagttggcgggcgcggattatttttgcggttttgtcggttggtt
gttattttcgggggttttggttgtcgatttacggggcgggttttcgagcggttttaagtttcggagttgggcgggggcgg
gagggagtttgggagaa-1914

Cloned sequence after bisulfate treatment and subsequent PCR

TTCTCCCAAACTCCCTCCCACCCCCACCCAACTCCAAAACTTAAAACCACTCAAAAACCCACCCCATAAA
TCAACAACCAAAACCCCCAAAAATAACAACCAACCAACAAAACCACAAAAATAATCCACACCCACCAACT
ACAAAATTACACAAAATATATCACACTAAC

-youryuer-

Hi,

the company you sent your samples to used the reverse primer to seuence your PCr product after cloning i.e. you have to take the reverse complement of the sequence you got from them and align it to the converted sequence. You'll see that they align very well and that you have the right product.

Bram

-Bram-

It doesn't work! There is still no significant similarity! I am wondering whether the PCR product is not my target product?

-youryuer-

Is methylnick there? Methylnick, could you help me analyze what is the reason? And your suggestions? Thanks a lot!

-youryuer-

It does work:

....|....| ....|....| ....|....| ....|....| ....|....| ....|....|
5 15 25 35 45 55
converted TTTTTTTGTT AGTGTGATAT ATTTCGCGTA ATTTCGTAGT TGGCGGGCGC GGATTATTTT
company -------GTT AGTGTGATAT ATTTTGTGTA ATTTTGTAGT TGGTGGGTGT GGATTATTTT

....|....| ....|....| ....|....| ....|....| ....|....| ....|....|
65 75 85 95 105 115
converted TGCGGTTTTG TCGGTTGGTT GTTATTTTCG GGGGTTTTGG TTGTCGATTT ACGGGGCGGG
company TGTGGTTTTG TTGGTTGGTT GTTATTTTTG GGGGTTTTGG TTGTTG-TTT ATGGGGTGGG

....|....| ....|....| ....|....| ....|....| ....|....| ....|..
125 135 145 155 165 175
converted TTTTCGAGCG GTTTTAAGTT TCGGAGTTGG GCGGGGGCGG GAGGGAGTTT GGGAGAA
company TTTTTGAGTG GTTTTAAGTT TTGGAGTTGG GTGGGGGTGG GAGGGAGTTT GGGAGAA

-Bram-

Dear Bram, you are very right and wise! I found the reason! Thank you very much!

It is not the problem with my sequences, to my surprise, it is the aligning tool bl2seq for blasting two sequences in the NCBI website that doesn't work. It is unbelievable! Just now, I aligned the whatever reverse&complemented&reverse complemented cloned sequence with the original&C-T conversion sequence, it still showed there is no significant similarity. Then I aligned again by using the tool Megalign in the software DNAStar, the index of similarity is 100%.

I feel happy you help me tackle this problem! Thanks again!

Best regards!

tony

-youryuer-

Dear Bram, 100% similarity occurs only when all the C including C in the CpGs are converted to T. It means the region I focus on is not methylated at all. I am curious how this can happen, since it has been reported in the literatures the region is regulated by DNA methylation with high extent methylated CpGs. Thanks very much!

-youryuer-

Hi

when you do pcr on genomic DNA you amplify dna from numerous cells. Not all the cells have the same methylation status therefore after pcr you see one band on a gel but it probably includes few different sequences - some that were methylated have "C"-s, unmethylated ones have "T"-s. Polymerase will preferentially amplify easy (T-rich) regions so after PCR you will have more amplicons reflecting unmethylated status of your DNA. Then you make ligation (still having more "T"- rich amplicons that will preferentially ligate with the vector, transform bacteria and what you analyze by sequencing is only one possible sequence that you have after your PCR.
Well i don't know if the above is right but it seems probable so if i were you i would analyze DNA from few more colonies after transformation.

and still it is possible you just don't have methylation there, hmm

BTW when you yuse blast2seq remember to unmark "filter"

anna

-gangut-

Dear anna, it is very kind of you for your clear answers! Yes, you are right! Each DNA molecule has the distincted methylated profile! Therefore, more clones are required for statistics. BTW, blseq2 does work when unmark filter. I was totally unware of function of "filter" before. Thanks for your suggestions!

tony

-youryuer-